Clinical Utility of Broad-Range Fungal PCR

BACKGROUND: Broad-range PCR is increasingly being used as a diagnostic tool in clinical microbiology. In this study we sought to determine the clinical utility of this test for detecting fungal pathogens. METHODS: A retrospective chart review was performed on all patients who had broad-range fungal PCR testing, performed at a reference laboratory, over a 25-month period. RESULTS: There were 32 broad-range fungal PCR tests performed on 25 patients during the study period, and six (19%) tests were positive for five (20%) patients. Of the patients with positive results, only in one (4%) case did the test result alter antifungal therapy: for this case, fungal cultures were not performed, and the only specimen available was formalin-fixed-paraffin-embedded tissue. Of the remaining patients with positive results, two had prior positive fungal cultures and were already initiated on antifungal therapy. In the last two patients, antifungal therapy was never initiated, in one case because the identified fungus was believed to be a contaminant and in the other because the infection resolved prior to the result. A positive or intermediate result for (1,3)-β-D-glucan was a common impetus for ordering broad-range fungal PCR in negative cases (n = 4, 27%). The clinical sensitivity and specificity, respectively, of fungal culture were 44% (n = 4/9) and 92% (n = 11/12) and of broad-range fungal PCR were 31% (n = 4/13) and 90% (n = 17/19), when defining true positives based on the clinical decision to treat. Patients for whom fungal organisms were seen in surgical pathology specimens were more likely to have a positive broad-range fungal PCR result (n = 1/2, 50%) than those for whom fungal organisms were not seen (n = 1/13, 8%) (P = 0.101). CONCLUSION: Fungal culture is more clinically sensitive and specific than broad-range fungal PCR. The results from broad-range PCR are more likely to be positive when fungal organisms are seen in surgical pathology specimens; however, based on our number of cases, these results are not statistically significant. Larger studies are needed to identify optimal situations for the utilization of broad-range fungal PCR. DISCLOSURES: All authors: No reported disclosures.